Dieter Kaiser wrote:
> Hello Ray,
>
> the code seems to be nearly complete. The first more systematically
> tests for the beta_incomplete function with float and bigfloat numbers
> work well.
>
> Again a small problem:
>
> The function bigfloat:expt does not work for a zero base:
>
> (%i23) :lisp (bigfloat:expt 0.0 (bigfloat:bigfloat 1.0));
> Maxima encountered a Lisp error:
> LOG: division by zero
> Automatically continuing.
> To reenable the Lisp debugger set *debugger-hook* to nil.
>
Oops. I think I've fixed the issues with expt. It still needs more
work though. Things like (expt -1b0 1b0) returns a complex number
instead of -1b0. But expt should return answers that are at least close
to the real answer.
>
> (%i28) :lisp (bigfloat:expt -1.0 (bigfloat:bigfloat 1.0));
> Maxima encountered a Lisp error:
> NO-APPLICABLE-METHOD: When calling #<STANDARD-GENERIC-FUNCTION
> BIGFLOAT::TWO-ARG-*> with arguments (+1.0b0 #C(0.0
> 3.141592653589793)), no method is applicable.
> Automatically continuing.
> To reenable the Lisp debugger set *debugger-hook* to nil.
>
>
I completely missed the case of bigfloat op cl:complex. I've fixed
these now.
I hope bigfloat works better now. Sorry for the trouble. If you find
more issues, just holler.
Ray